James Geoffrey Chase

Publications (467) View all

• Source

  Available from: Min-Ho Chey

  Article: Innovative seismic retrofitting strategy of added stories isolation system

  Min-Ho Chey, J Geoffrey Chase, John B Mander, Athol J Carr
  [show abstract] [hide abstract]
  ABSTRACT: The seismic performance of "added stories isolation" (ASI) systems are investigated for 12-story moment resisting frames. The newly added and isolated upper stories on the top of the existing structure are rolled to act as a large tuned mass damper (TMD) to overcome the limitation of the size of tuned mass, resulting to "12 + 2" and "12 + 4" stories building configurations. The isolation layer, as a core design strategy, is optimally designed based on optimal TMD design principle, entailing the insertion of passive flexible laminated rubber bearings to segregate two or four upper stories from a conventionally constructed lower superstructure system. Statistical performance metrics are presented for 30 earthquake records from the 3 suites of the SAC project. Time history analyses are used to compute various response performances and reduction factors across a wide range of seismic hazard intensities. Results show that ASI systems can effectively manage seismic response for multi-degree-of freedom (MDOF) systems across a broader range of ground motions without requiring burdensome extra mass. Specific results include the identification of differences in the number of added story by which the suggested isolation systems remove energy.
  The IES Journal Part A Civil & Structural Engineering 03/2013; 7(1):13-23.
• Source

  Available from: Irene Hudson

  Dataset: Density Estimation and Wavelet Thresholding via Bayesian Methods: A Wavelet Probability Band and Related Metrics Approach to Assess Agitation and Sedation in ICU Patients

  In Kang, Irene Hudson, Andrew Rudge, J Geoffrey Chase
  [show abstract] [hide abstract]
  ABSTRACT: A wave is usually defined as an oscillating function that is localized in both time and frequency. A wavelet is a “small wave”, which has its energy concentrated in time providing a tool for the analysis of transient, non-stationary, or time-varying phenomena [1, 2]. Wavelets have the ability to allow simultaneous time and frequency analysis via a flexible mathematical foundation. Wavelets are well suited to the analysis of transient signals in particular. The localizing property of wavelets allows a wavelet expansion of a transient component on an orthogonal basis to be modelled using a small number of wavelet coefficients using a low pass filter [3]. This wavelet paradigm has been applied in a wide range of fields, such as signal processing, data compression and image analysis [4 -10]. Typically agitation-sedation cycling in critically ill patients involves oscillations between states of agitation and over-sedation, which is detrimental to patient health, and increases hospital length of stay [11-14]. The goal of the research specifically in reference [14] was to develop a physiologically representative model that captures the fundamental dynamics of the agitationsedation system. The resulting model can serve as a platform to develop and test semiautomated sedation management controllers that offer the potential of improved agitation management and reduce length of stay in the intensive care unit (ICU). A minimal differential equation model to predict or simulate each patient’s agitation-sedation status over time was presented in [14] for 37 ICU patients, and was shown to capture patient A-S dynamics. Current agitation management methods rely on subjective agitation assessment and an appropriate sedation input response from recorded at bedside agitation scales [15, - 19]. The carers then select an appropriate infusion rate based upon their evaluation of these scales, their experience and intuition [20]. This process is depicted in Figure 1 (see [14]). Recently a more refined A-S model, which utilised kernel regression with an Epanechnikov kernel and better captured the fundamental agitation-sedation (A-S) dynamics was formulated [12, 13]. A secondary aim of this chapter is to test the feasibility of wavelet statistics to help distinguish between patients whose simulated A-S profiles were “close” to their mean profile versus those for whom this was not the case (i.e. their simulated profiles are not “close” to their actual recorded profiles). This chapter builds on a preliminary study [21] to assess wavelet signatures for modelling ICU agitation-sedation profiles, so as to, as in this chapter, evaluate “closeness” or “discrimination” of simulated versus actual A-S profiles with respect to wavelet scales - as recently analysed using DWT and wavelet correlation methods in [29] (see also [22]-[24]). The recent work of Kang et al. [29] investigated the use of DWT signatures and statistics on the simulated profiles derived in [12] and [13], to test for commonality across patients, in terms of wavelet (cross) correlations. Another earlier application of this approach was the study of historical Australian flowering time series [22], where it was established that wavelets add credibility to the use of phenological records to detect climate change. This study was also recently expanded and reported by Hudson et al. [23, 24] (see also references [25-28]). The density function is very important in statistics and data analysis. A variety of approaches to density estimation exist. Indeed the density estimation problem has a long history and many solutions [30, 31, 32]. A large body of existing literature on nonparametric statistics is devoted to the theory and practice of density estimation [32-36]. The local character of wavelet functions is the basis for their inherent advantage over projection estimators – specifically that wavelets are straightforward and well localized in both space and frequency. The relevant estimation methods belong to the class of so-called projection estimators, as introduced by [36] or their non-linear modifications. Section 3 traces, in brief, the development of some basic methods used in density estimation. We then link these and apply wavelet methods for (density) function estimation to the ICU data of [29]. In this chapter the density is estimated using wavelet shrinkage methods, as based on Bayesian methods. Specifically the minimax estimator is used to obtain a patient specific wavelet tracking coverage index (WTCI). All values of the WTCI are obtained using Bayesian wavelet thresholding, and are shown to differentiate between poor versus good tracking. A Bayesian approach is also suggested in this chapter by which to assess a parametric A-S model – this by constructing a wavelet probability band (WPB) for the proposed model and then checking how much the nonparametric regression curve lies within the band. The wavelet probability band (WPB) is shown to provide a useful tool to measure the comparability between the patient’s simulated and recorded profiles. Moreover, the density profile is then successfully used to define and compute two numerical measures, namely the average normalized wavelet density (ANWD) and relative average normalized wavelet density (RANWD) – both measures of agreement between the recorded infusion rate and simulated infusion rate. Our WPB method is shown to be a good tool for detecting regions where the simulated infusion rate (model) performs poorly, thus providing ways to help improve and distil the deterministic A-S model. The so-called Wavelet Time Coverage Index (WTCI) developed is analogous to the metrics based on a kernel based probability band of [13, 14]. The research in [29] and that formulated in this chapter have successfully developed novel quantitative measures based on wavelets for the analysis of A-S dynamics. http://dx.doi.org/10.5772/52434
• Article: Preface.

  J Geoffrey Chase, Ewart Carson
  Computer methods and programs in biomedicine 02/2013; 109(2):113-5. · 1.14 Impact Factor
• Article: Analysis of different model-based approaches for estimating dFRC for real-time application.

  Erwin J van Drunen, J Geoffrey Chase, Yeong Shiong Chiew, Geoffrey M Shaw, Thomas Desaive
  [show abstract] [hide abstract]
  ABSTRACT: BACKGROUND: Acute Respiratory Distress Syndrome (ARDS) is characterized by inflammation, filling of the lung with fluid and the collapse of lung units. Mechanical ventilation (MV) is used to treat ARDS using positive end expiratory pressure (PEEP) to recruit and retain lung units, thus increasing pulmonary volume and dynamic functional residual capacity (dFRC) at the end of expiration. However, simple, non-invasive methods to estimate dFRC do not exist. METHODS: Four model-based methods for estimating dFRC are compared based on their performance on two separate clinical data cohorts. The methods are derived from either stress-strain theory or a single compartment lung model, and use commonly controlled or measured parameters (lung compliance, plateau airway pressure, pressure-volume (PV) data). Population constants are determined for the stress-strain approach, which is implemented using data at both single and multiple PEEP levels. Estimated values are compared to clinically measured values to assess the reliability of each method for each cohort individually and combined. RESULTS: The stress-strain multiple breath (at multiple PEEP levels) method produced an overall correlation coefficient R2 = 0.966. The stress-strain single breath method produced R2 = 0.530. The single compartment single breath method produced R2 = 0.415. A combined method at single and multiple PEEP levels produced R2 = 0.963. CONCLUSIONS: The results suggest that model-based, single breath and non-invasive approaches to estimating dFRC may be viable in a clinical scenario, ensuring no interruption to MV. The models provide a means of estimating dFRC at any PEEP level. However, model limitations and large estimation errors limit the use of the methods at very low PEEP.
  BioMedical Engineering OnLine 01/2013; 12(1):9. · 1.40 Impact Factor
• Article: Daily evolution of insulin sensitivity variability with respect to diagnosis in the critically ill.

  Tamás Ferenci, Balázs Benyó, Levente Kovács, Liam Fisk, Geoffrey M Shaw, J Geoffrey Chase
  [show abstract] [hide abstract]
  ABSTRACT: This study examines the likelihood and evolution of overall and hypoglycemia-inducing variability of insulin sensitivity in ICU patients based on diagnosis and day of stay. An analysis of model-based insulin sensitivity for [Formula: see text] patients in a medical ICU (Christchurch, New Zealand). Two metrics are defined to measure the variability of a patient's insulin sensitivity relative to predictions of a stochastic model created from the same data for all patients over all days of stay. The first selectively captures large increases related to the risk of hypoglycemia. The second captures overall variability. Distributions of per-patient variability scores were evaluated over different ICU days of stay and for different diagnosis groups based on APACHE III: operative and non-operative cardiac, gastric, all other. Linear and generalized linear mixed effects models assess the statistical significance of differences between groups and over days. Variability defined by the two metrics was not substantially different. Variability was highest on day 1, and decreased over time ([Formula: see text]) in every diagnosis group. There were significant differences between some diagnosis groups: non-operative gastric patients were the least variable, while cardiac (operative and non-operative) patients exhibited the highest variability. This study characterizes the variability and evolution of insulin sensitivity in critically ill patients, and may help inform the clinical management of metabolic dysfunction in critical care.
  PLoS ONE 01/2013; 8(2):e57119. · 4.09 Impact Factor